fra81

[fra81]

Hi,

can you tell us about:

– Fedora version
– Plasma version
– Graphics card and drivers

[fra81]

So now we know that VirtualGL works when used through vglrun, the only problem is with the environment, the reason of which is still unclear. Do you confirm that no system script overwrites, unsets, or affects in any way the LD_PRELOAD variable?

LD_PRELOAD=/usr/NX/lib/libnxegl.so

Please note that the libnxegl library has a different purpose (it’s used for capturing the screen from the physical display when Wayland is in use), it should only be set in the physical desktop and it’s not useful in virtual desktops. In virtual desktops, when VirtualGL is enabled, LD_PRELOAD should look like this:

LD_PRELOAD=/usr/NX/scripts/vgl/libdlfaker.so:/usr/NX/scripts/vgl/libvglfaker.so

env XDG_DATA_DIRS=/usr/local/share:/usr/share:/var/lib/snapd/desktop /usr/NX/scripts/vgl/vglrun -d /dev/dri/card0 /usr/bin/dbus-launch –sh-syntax –exit-with-session /usr/bin/startplasma-x11

This is definitely possible and, maybe, it could help with working around the environment problem. Note that you will have to disable VirtualGL in the server configuration (/usr/NX/bin/nxserver --virtualgl no) after putting the vglrun command in the DefaultDesktopCommand key.

However, I would recommend a slightly different default desktop command from the one you posted:

env XDG_DATA_DIRS=/usr/local/share:/usr/share:/var/lib/snapd/desktop /usr/bin/dbus-launch –sh-syntax –exit-with-session /usr/NX/scripts/vgl/vglrun -d /dev/dri/card0 /usr/bin/startplasma-x11

[fra81]

Hi @boospy

[VGL] ERROR: Could not open display :0.

Did you get that in the physical desktop or in the NoMachine virtual desktop? Can you run ‘env | grep DISPLAY’ in the physical desktop? If you find out that the display number is, e.g., :1 instead of :0, you can adjust the vglrun command in this way:

/usr/NX/scripts/vgl/vglrun -d :1 glxinfo | grep -i “renderer\|vendor”

Or, even better, you can instruct VirtualGL to use the GPU directly without passing for the X server. This is a new feature of VirtualGL. So, if the path to your video device is e.g. /dev/dri/card0, you can use this command for vglrun:

/usr/NX/scripts/vgl/vglrun -d /dev/dri/card0 glxinfo | grep -i “renderer\|vendor”

Using the -d option is equivalent to setting the VGL_DISPLAY variable in the environment (e.g. VGL_DISPLAY=:0 or VGL_DISPLAY=/dev/dri/card0).

What’s the output of the vglrun command after these adjustments?

The variable is displayed/read out correctly when I log in on the physical screen. Unfortunately not via the virtual Nomachine session.

This is indeed strange, I suspect that something in your user profile is unsetting the LD_PRELOAD variable (I see that it is correctly set by the server at the moment it launches the desktop environment in the virtual desktop). Could you check if this is the case in your user profile files (.profile, .bashrc, .bash_profile, /etc/profile, /etc/profile.d/*, etc)? Or are you aware of anything that might be interfering?

[fra81]

Hi Chaoimin,

I don’t have an explanation why restarting SDDM helps and restarting the NoMachine server doesn’t, since the nature of the problem is not clear yet, and we’re still unable to reproduce it on our labs. This could also indicate an issue in KDE Plasma rather than NoMachine.

Can you please try to press a key on the physical keyboard of the laptop you use as a server when you reproduce the issue, and check if input events through the NoMachine session start working from that moment on?

Did you restart the NoMachine server (/usr/NX/bin/nxserver --restart) after adding the DisplayServerExtraOptions “-eventdelay 0” option in the node.cfg file? Would you try again if you didn’t?

Is there any screen lock app running on the server when the issue is being reproduced?

[fra81]

Hi,

is that the home dir of the same user who owns the desktop?

What is the desktop environment (Gnome, KDE…) in use and version?

Does ‘/usr/NX/bin/nxserver –eglcapture no’ help to solve the crashes? I forgot to say that you need to logout and login again, or reboot the host, to make this change effective.

[fra81]

@psantucc,

while the test package is being prepared, can you tell about the GPU model and drivers version?

It could also be useful to know about the GL/EGL version. To do that, you can run these commands in a terminal on the server:

glxinfo | grep "OpenGL version"
eglinfo | grep "EGL version"

Before that, you may need to install ‘mesa-utils’ package providing the glxinfo command.

Please also try to add this line into the /usr/NX/etc/node.cfg file on the server:

WaylandModes drm,egl,compositor

Then restart the NoMachine server. This will make an alternative capture method (“drm”) be used before the default one (“egl”). Does this workaround help?

[fra81]

Hi Detlef,

Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x0000000000000000 in ?? ()
[Current thread is 1 (Thread 0x78f22be006c0 (LWP 461301))]
(gdb)

When you get to that point in gdb, please type the ‘bt’ command and show us the output.

“Enable hardware accelerated decoding” false or disabled do not change anything.

Did you change the key to ‘disabled’ after quitting the NoMachine player? Here are more details on how to set it: https://kb.nomachine.com/AR07U01202. Please make sure it is set to ‘disabled’ rather than ‘false’.

[fra81]

Hi,

this problem has never been experienced in our labs or even reported. As a quick try, please add this line to the
‘/usr/NX/etc/node.cfg’ file on the host (server) machine:

DisplayServerExtraOptions "-eventdelay 0"

Then restart the NoMachine server. Does it help?

 

[fra81]

Unfortunately we can’t reproduce even by following those steps, but we will try on another system. In the meanwhile, are you using Wayland as a display server on that Fedora? Can you try to disable it (switch to Xorg)?

[fra81]

Hi,

we can’t reproduce this issue in our labs, so we would need some more info from you.

1) You say you weren’t experiencing the problem before updating. Do you mean updating the client or the server?

2) As I understand, both clients run on ARM architecture. Would it be possible to try a different client? And, if using the same two clients, do you have the possibility to connect to a different server machine and check whether the issue is present also there?

3) What is the model of your Mac?

4) Can you please try to rename this library on the server:

sudo mv /Applications/NoMachine.app/Contents/Frameworks/lib/libnxsck.dylib /Applications/NoMachine.app/Contents/Frameworks/lib/libnxsck.dylib-bak

and restart the NoMachine server? Does it help?

5) May you send us the logs from the server? See https://kb.nomachine.com/DT07S00243. You can send them to forum[at]nomachine[dot]com.

[fra81]

Hi,

this is the first time a similar issue has been reported, so we want to try and reproduce it in our lab. What is the Linux distribution and version you are using? What desktop environment?

Please also compress the nxegl log directory ($HOME/.nx/nxegl) and send us. You can attach the archive here or send to forum[at]nomachine[dot]com.

Is this library needed for client-only use of NoMachine? (e.g. I won’t ever be connecting to this Linux machine, only from this Linux machine to other Windows/macOSes).

It isn’t. You can disable the loading of this library with this command:

sudo /usr/NX/bin/nxserver --eglcapture no

[fra81]

Hi akevinge,

those logs are not relevant indeed.

In virtual desktops software rendering is normally used and it seems like the software renderer provided by the video drivers is broken. Did you install proprietary drivers on the server? Can you try to update or uninstall them? Or, if not installed already, can you try to install them?

Other possible workarounds are:

– enabling VirtualGL for 3D rendering support (https://kb.nomachine.com/AR05P00982)

– using a different desktop environment (https://kb.nomachine.com/AR04K00667)

[fra81]

Hi,

so please try this other workaround. Open the Keyboard Viewer on the Mac, then click on the three dots and select Customize. Click on Add panel and select the ‘ANSI (Large)’ panel, so that you will have the numeric key pad on the Keyboard Viewer, including the NumLock button (normally labelled with an X sign). Now you can activate the NumLock locally on the Keyboard Viewer and from now on you should be able to use the numeric key pad of the physical keyboard.

[fra81]

Hi,

it seems that the X11 agent is not detecting correctly the actual XKB rules settings. In fact, xfree86 rules are used in the xrdp session, which are obsolete, instead of the evdev rules, that are now generally used. We will fix that in one of the next updates, but for now, as a workaround, running this command in the xrdp session should do the trick:

setxkbmap -rules evdev

[fra81]

Hi, would you try this workaround?

1.$ cd /usr/NX/lib
2. $ sudo mv libvpl.so libvpl.so.ori
3. $ find / -name "libvpl.so*"
4a. If ‘libvpl.so’ (the exact name, without suffixes) is found in the system by the previous command, then skip to step 7 to restart the server
4b. If the libvpl library is found with a slightly different name like ‘libvpl.so.2′ or similar, then proceed to step 5 to create a symlink
5. $ cd <directory_where_is_libvpl.so.2>
6. $ sudo ln -s libvpl.so.2 libvpl.so
7. $ /usr/NX/bin/nxserver --restart

[Author]

Posts